Genetic and Microbial Influences on the Development of Obesity in Children

<p>NON-TECHNICAL SUMMARY:<br/>Childhood obesity is a nutrition-related disease with multiple underlying etiologies. The gut microbiota is thought to be a contributor in the development of obesity by fermentation of non-digestible polysaccharides to short chain fatty acids (SCFA), which increases host capacity for energy harvest and storage. Several genes encoding SCFA receptors and transporters, as well as other host responders to gut microbiota have been described. However, the collective impact of common genetic variations (single nucleotide polymorphisms [SNP]) in these genes on obesity phenotypes has yet to be examined in humans. Our hypothesis is that genetic variation in SCFA recognition pathways are related to excess weight gain and microbial distinct profiles are associated with overweight and obesity in children. This proposal will also use a unique approach in
studying the relationship between gut microbiota and obesity by evaluating both traditional anthropometric measurements and percent body fat measured by dual energy x-ray absorptiometry (DXA). This research proposal aligns with the challenge area regarding improving nutrition and ending childhood obesity. The innovative approach here may reveal novel biomarkers for obesity susceptibility at an early age. Analysis of genetic factors will provide needed information for the development of early detection methods including genetic screening for obesity risk. Additionally, an understanding of key microbial players in the gut of children with normal or high BMI and percent body fat will pave the way for therapeutics designed to achieve and maintain optimal gut health.

<p>APPROACH:<br/>Our approach will consist of building upon our research thus far by increasing our sample size in both the gut microbiota and genetic components of the project. This will be accomplished by utilizing saliva collection that has occurred through the University of Michigan's STRONG Kids cohort (Ann Arbor, MI; n=150), which will be added to the existing cohort of children from Illinois' STRONG Kids Program, for DNA extraction and genetic association with obesity. Addition of these samples will strengthen our ability to conduct genetic association studies with obesity phenotypes. Post-hoc power analysis has indicated that despite our relatively small sample size, the effect size (d=1.7) was sufficient to detect differences in bacterial quantities between ow/ob (n=6) and lean (n=12) individuals (1-?=0.89). However, to detect differences where the
effect size is smaller, we will include additional cross-sectional sampling (n=20). We will recruit children (4-5 years old) from Central Illinois to collect stool, breath, and perform DXA to include with the existing data set. The protocol for this recruitment is already in place and has received IRB approval.
<p>The following methods will be performed: First, DNA will be extracted from stool using the protocol used by Nava and et al. Quantitative RT-PCR will be performed to quantify levels of bacteria in Clostridium cluster XIVa, Clostridium cluster IV, Lactobacillus, Bifidobacterium, and Bacteroides-Prevotella. Second, DXA scans will be performed using Hologic Discovery A bone densitometer (Bedford, MA), and percent body fat measurements will be obtained with Hologic software (version 12.7.3.1, QDR-Discovery A). Third, breath samples will be collected using a technique developed by our
laboratory (in press), and exhaled hydrogen, methane, and carbon dioxide will be measured by the BreathAnalyzer SC by Quintron Instrument Company (Milwaukee, WI). Exhaled methane is a reflection of gut microbial metabolism, and we will test whether associations described between elevated breath methane and obesity can be detected in children. All methods employed in this study are feasible, as equipment and reagents are readily available through the laboratories of the primary mentor and co-advisor. Statistical analysis will be performed using SAS 9.3 using T-test, Spearman partial correlation, stepwise regression, and principal component analysis. Methods for testing individual and cumulative effects of genetic variants on obesity will include the following: Human genomic DNA will be extracted from saliva following standard protocols already established. From height and weight data,
BMI, BMI percentile, and anthropometric Z-scores, including weight-for-age, height-for-age, weight-for-height, and BMI-6for-age Z-scores will be calculated by using the standard SAS Program from the Centers for Disease Control and Prevention. SNP selection will be performed using two approaches: Initially, the NCBI database will be searched for functional and tagged SNPs in the following genes of interest (FFAR2, FFAR3, ANGPTL4, SLC16A1, SLC5A8, SLC5A12, TLR4, LPL, PYY, leptin, and GLP-1) with a minor allele frequency of 0.2 or greater. Many of these genes have been previously described as playing a role in the recognition of gut microbiota and SCFAs. Next, selected SNPs will be evaluated for allele frequency in a STRONG Kids case-control subset already genotyped using the Human Omni Express Illumina array. Markers showing heterozygosity (>0.5) will be further genotyped in the full
cohort of individuals using conventional genotyping assays. General linear models will be used to test for genetic associations with obesity phenotypes including BMI, BMIPCT, and anthropometric z-scores with age, sex, and breastfeeding duration as covariates. Additionally, GPS will be generated using publically available software (i.e. Plink) according to an additive risk model.